Short-term effects of 90/90 breathing with ball and balloon on core stability

Master Thesis

HALMSTAD

UNIVERSITY

Master's Programme in Exercise Biomedicine - Human

Performance, 60 credits

Short-term effects of 90/90 breathing with

ball and balloon on core stability

Biomedicine, 30 credits

Halmstad 2018-05-29

Lukas Alverdes

Short-term effects of 90/90 breathing with

ball and balloon on core stability

Lukas Alverdes

2018-05-29

Master Thesis 30 credits in Exercise Science – Human Performance

Halmstad University

School of Business, Engineering and Science

Thesis supervisor: Sofia Ryman Augustsson

Thesis examiner: Åsa Andersson

Acknowledgements

First of all, I would like to thank my supervisors in Halmstad, Sofia Ryman Augustsson, as well as Daniel Gärtner in Munich for supporting me along my way of writing this thesis. Moreover, I want to thank Hanna and the whole team of the university gym in the Halmstad University for supplying me with the tools that I needed to conduct my tests. Thanks also go to Lina Lundgren for all the help with the laboratory and its equipment. Last I wanted to thank my lector and good friend Arne for helping me with the final touches, my friends Lars and Nicolai for the help and advices along the conception and writing of this thesis and my friend Boris for the support during this project.

Abstract

Background Breathing is a life preserving mechanism that can influence muscles of the core

and its stabilizing mechanisms, especially by the function of the diaphragm and intra-abdominal pressure (IAP) build-up. The 90/90 bridge with ball and balloon (90/90 breathing) is one technique doing so, thereby affecting the core and core stability (CS). Both have been shown to influence injury, and in some studies performance, and are therefore deemed important. In the Functional Training branch exercises that influence CS are used as core activations in the warm-up to increase performance in the short-term, but scientific proof for that is lacking.

Objective The aim of this study was therefore to investigate if a core activation in the form of

the 90/90 breathing can increase the short-term CS. Methods To test this an intervention trial was designed where the subjects were divided into a control group (CG) and a breathing group (BG). Three CS-tests were done to assess the CS at two times, Pre and Post. The double-leg-lowering (DLL), the unilateral-hip-bridge (UHB) and the single-leg-stand (SLS). The BG did the 90/90 breathing in between Pre and Post, whereas the CG did nothing. The data was checked for group differences at Pre and Post as well as the difference in the performance change from Pre to Post between groups using Independent t-test and Mann-Whitney U test. Improvements from Pre to Post within groups were calculated with Pared Samples t-test and Wilcoxon tests.

Results No consistent effect of the intervention was found. The DLL showed the most positive

results with a performance improvement in the BG and a greater performance change for the BG than for the CG. The UHB showed mixed results with a better performance at Post for the BG in both legs but only an improvement for the non-dominant leg in the BG. The SLS showed no improvement for the BG in any test. Conclusion The inconsistent results show no general positive effect of the 90/90 breathing on CS. However, the positive effects in the DLL make a position and task specific effect of the 90/90 breathing on CS possible. Practitioners and coaches should consider this task specificity when planning warm-ups. Future research should also choose CS tests and training exercises more task specific to the studied objectives to obtain more distinct results. More research on the short-term effects of CS interventions is needed for a clearer understanding of the subject.

List of Abbreviations

90/90 breathing 90/90 bridge with ball and balloon

AB Abdominal bracing

APA Anticipatory postural adjustment

BG Breathing Group

CG Control Group

COM Center of mass

COP Center of pressure

COPD Chronic obstructive pulmonary disease

CS Core stability

DB Dysfunctional breathing

DL Dominant leg

DLL Double leg lowering

EMG Electromyography

FB Functional breathing

FT Functional Training

HLT High-low-test

IAP Intra-abdominal pressure

ICC Intra-class correlation coefficient

LBP Lower back pain

LRET Lateral rib expansion test

N Newton

NDL Non-dominant leg

PDB Partly dysfunctional breathing

SLS Single leg stand

SSPW Sport sessions per week

TSDS Time spent doing sports per week

UHB Unilateral hip bridge

YTE Years of training experience

Table of Content

2.2 Effects of functional and dysfunctional breathing ... 4

2.3 Breathing and stability ... 5

2.3.1 The role of the diaphragm in stability ... 5

2.3.2 The effect of 90/90 breathing on stability ... 6

2.4 Anatomical and functional definition of the core ... 8

2.5 Core stability definition and components ... 9

2.6 Measuring Core stability ... 10

2.7 Core stability and its influence on injury and performance ... 11

2.7.1 Core stability and injury ... 11

2.7.2 Core stability and performance ... 12

2.8 Functional Training ... 14 2.9 Aim ... 15 3. Methods ... 15 3.1 Subjects ... 15 3.2 Experimental Design ... 16 3.3 Testing procedure ... 17 3.4 Breathing intervention ... 20

3.5 Ethical and Social Considerations ... 21

3.6 Statistical Analysis ... 21

4. Results... 22

4.1 Descriptive statistics ... 22

4.2 Main effect calculations ... 23

4.2.1 Group difference at the first and second test round ... 23

4.2.2 Performance change within groups from the first to second test round ... 24

4.2.3 Difference in the performance change between groups ... 25

4.3 Secondary calculations ... 26

5. Discussion ... 31

5.1 Results Discussion ... 32

5.1.1 Inconsistency in the results ... 32

5.1.3 No effect of the 90/90 breathing on core stability ... 35

5.1.4 No impact of gender, breathing pattern and sportiveness on core stability ... 36

5.2 Methods discussion ... 37

5.2.1 Appropriateness of the tests ... 37

5.2.2 Appropriateness of the intervention ... 38

5.2.2 Appropriateness of the study design ... 39

5.2.3 Reliability and Validity issues ... 39

6. Conclusion ... 40

6.1 Practical implications ... 41

6.2 Recommendations for future research ... 41

7. References ... 42

1

1. Introduction

In this Master Thesis it was investigated whether a breathing technique could have a positive influence on core stability (CS) and improve it in the short-term. This is of importance since breathing is a basic, life preserving mechanism of the body with more impact on non-respiratory mechanisms than many people are aware of. Breathing is actively used in the practical field, especially by traditional sportive or meditative activities like Yoga, and recently more by upcoming branches of the fitness industry like Functional Training. There breathing is mainly a technique for relaxation and working against disbalances (Boyle, Olinick & Lewis, 2010). But in recent years it has also become a focus for increasing stability during exercises and overall sports performance by using different breathing techniques (Bradley & Esformes, 2014; Tayashiki, Mizuno, Kanehisa, & Miyamoto, 2017; Illi, Held, Frank, & Spengler, 2012; Hodges & Gandevia, 2000b). One of the newer breathing techniques that gained more popularity recently is the 90/90 bridge with ball and balloon, that was originally designed to work against disbalances and pain in the upper body and torso (Boyle et al., 2010). However, this type of breathing also works on and activates a lot of different muscles in the core (Goldman, Lehr, Millar & Silver, 1987; Kim & Lee, 2017). In the Functional Training these muscles are seen as an important part of the warm-up, where exercises like plank variations are designed to activate them. This is supposed to increase the core stability for the strength or dynamic exercises to come. However, this effect has barely been scientifically proven yet and the question stands if a short-term improvement in core stability with exercises that work on the core is really possible. This led to the question if an activation of the core muscles, by doing the 90/90 bridge with ball and balloon, could increase the short-term core stability. Answering this question has implications for the Functional Training (FT) field by being able to design the warm-up part of training sessions more efficiently. It could also benefit the general-public especially for people with lower back pain (LBP), since core stability as well as breathing have been shown to play a role in its rehabilitation and prevention (Boyle et al., 2010; Chang, Lin, & Lai, 2015). Taken together, the results of this study could provide helpful information for people suffering from conditions like LBP and the FT community for a more efficient warm-up, as well as for the scientific community concerning CS research.

2

2. Background

2.1 Breathing mechanism

2.1.1 Functional Breathing

The lungs are passive extensible organs; therefore, they need the respiratory muscles to inflate, and in some situations to deflate. The muscles responsible for inspiration are the diaphragm, external intercostals, parasternal, sternomastoid and scalene muscles (Ratnovsky, Elad, & Halpern, 2008). The muscles responsible for expiration are the internal intercostal, rectus abdominis, external and internal oblique and transversus abdominis muscles (Ratnovsky et al., 2008). They are all located around the torso (see Figure 1).

Figure 1: Position of the respiratory muscles in the torso (Ratnovsky et al., 2008).

In this group of breathing muscles, the diaphragm has a special role. Amongst other reasons because it is the main inspiratory muscle during correct breathing, especially while being at rest (Sharma, 2012). All the respiratory muscles are important for the function of the lungs, but they have different times and situations when they are supposed to be active. During quiet breathing there is supposed to be synchronized motion of the lower and upper rib cage as well as the abdomen. All this motion is mainly produced by the diaphragm with help of the scalene, upper parasternal and intercostal muscles (Sharma, 2012; Bradley & Esformes, 2014). The magnitude of help of the accessory respiratory muscles differs between studies. Some show no activity during quiet breathing (Costa, Vitti & de Oliveira Tosello, 1997), while others show activity during quiet breathing (Perri & Halford, 2004; Ratnovsky, Zaretsky, Shiner & Elad, 2003). Therefore, normal breathing is also called diaphragmatic breathing. It is characterized by a

3

three-dimensional expansion of the lower ribcage in combination with movement of the abdominal wall (see Figure 2) as main mechanisms of breathing (Bradley & Esformes, 2014; Courtney, 2009). During normal breathing, only the inspiratory muscles are active, and the expiration happens automatically by the elastic recoil of the lung (Ratnovsky et al., 2008).

Figure 2: Representation of functional breathing with diaphragm activity and lateral rib expansion (Rakhimov, n.d.)

This breathing is sufficient during rest, but there are situations when the oxygen need of the body is increased. Most commonly this happens during exercise, when the greater number of active muscles and muscle fibers consume more energy and thereby need bigger amounts of oxygen to make the energy production possible (Your lungs and exercise, 2016). To make this increased airflow possible the diaphragm must increase its activity and get help from the accessory muscles (Sharma, 2012). Studies show their activity goes up when inspiratory effort increases (Ratnovsky et al., 2003), which leads to the increase in chest movement during breathing that is commonly seen in people engaging in strenuous effort.

2.1.2 Dysfunctional Breathing

This chest dominant breathing is known as abnormal breathing if seen during rest, when the increased need for oxygen is not given (Perri & Halford, 2004). Its characteristics are an increased upper rib cage motion compared to the motion of the lower rib cage and the abdominal area (Bradley & Esformes, 2014). This decreased abdominal motion in comparison to the increased thoracic motion is not only a sign of thoracic breathing produced by the accessory

4

muscles, it might also be an indicator of poor diaphragm activity (Bradley & Esformes, 2014). In this faulty pattern muscles that are usually no breathing muscles, like the pectoralis major and minor, the latissimus dorsi, serratus anterior and trapezius, become accessory respiratory muscles (Perri & Halford, 2004). This faulty pattern increases the risk of their overuse (Courtney, 2009), possibly reducing their function in other tasks.

The prevalence for dysfunctional breathing differs between studies. Some report that between 5-11% of the people in the healthy population suffer from dysfunctional breathing (Courtney, 2009). Others show numbers of 56.4% dysfunctional breathing during rest and 75% during deep breaths in the average population (Perri & Halford, 2004). Considered that the average person takes around 21.000 breaths per day (Courtney, 2009), a dysfunction in every one of them has the potential of detrimental effects on the body and the heavily overused accessory respiratory muscles.

2.2 Effects of functional and dysfunctional breathing

If dysfunctional breathing patterns have become a habit several negative short- and long-term effects can develop out of that. The study done by Bradley and Esformes (2014) shows that movement efficiency might be affected by dysfunctional breathing, which could be explained by the fact that “breathing is one of the most basic patterns directly related to human movement” Cavvaggioni, Ongaro, Zannin, Iaia and Alberti (2015, p. 1). Dysfunctional breathing can also have negative effects on the posture due to an overload in the accessory breathing muscles (Courtney, 2009). This overload can lead to several adverse effects with symptoms like headache (Hruska, 1997) and neck pain (Perri & Halford, 2004; Bradley & Esformes, 2014; Hruska, 1997). Another common disorder in the general public, that is strongly influenced by dysfunctional breathing, is lower back pain (LBP). Studies show that a dysfunction in breathing muscles like the diaphragm, the transversus abdominis or the pelvic floor, can increase the chance to develop conditions like LBP or to get injured (Bradley & Esformes, 2014; Courtney, 2009; Roussel et al., 2009; Kolář et al., 2012). This is because besides their role in breathing they are important for the motor control and postural support of the body and especially of the spine. These negative postural effects in combination with the impaired motor control can also lead to a decrease in stability (Bradley & Esformes, 2014).

As made clear now it can have several negative effects on the body if the breathing pattern is dysfunctional. However, proper breathing patterns in form of breathing techniques can be used for actively working against problems. Probably the most popular use for breathing as a

5

therapeutic measure is against stress (Courtney, 2009; Jerath, Edry, Barnes & Jerath, 2006). Evidence also points towards an effect of breathing therapies in diseases like asthma, heart disease, anxiety, depression (Courtney, 2009), obstructive pulmonary disease (COPD), sciatica or thoracic outlet syndrome (Boyle et al., 2010). Correct breathing can also work against pain, especially in the neck and lower back area (Boyle et al., 2010) and increase overall stability and posture (Obayashi, Urabe, Yamanaka & Okuma, 2012). Because proper breathing is not automatic (Nelson, 2012), training of functional breathing is necessary for those who have lost it and suffer under the bad effects of dysfunctional breathing. Moreover, breathing exercises can enhance performance (Levine, 2002; Illi et al., 2012; Enright, Unnithan, Heward, Withnall & Davies, 2006). However, effects of breathing on performance and stability have mostly been researched with longer training regimes. The short-term effects of breathing interventions in these areas have hardly been studied so far.

2.3 Breathing and stability

2.3.1 The role of the diaphragm in stability

Decreased diaphragmatic activity, as evident in dysfunctional breathing, could be detrimental for performance since the diaphragm plays an important role in the stabilization of the lumbar spine area. Hodges, Butler, McKenzie and Gandevia (1997) found the first evidence for the postural role of the diaphragm by proving that it activates in an anticipatory postural adjustment (APA) fashion before the onset of muscles producing limb movement. Thereby the diaphragm contracts to help the abdominals and the pelvic floor to increase the intra-abdominal pressure (IAP), as can be seen in Figure 3. This activation occurs irrespective of the respiratory phase, proving that the diaphragm exerts a double role for respiration and postural stability (Hodges et al., 1997). Other studies confirmed this finding directly and indirectly, bringing further evidence for the importance the diaphragm plays in generating IAP (Kolář et al., 2012; Hodges & Gandevia, 2000a, 2000b; Hagins & Lamberg, 2011; Kawabata, Shima & Nishizono, 2014; Shirley, Hodges, Eriksson & Gandevia, 2003).

The generation of IAP is the main stability mechanism of the diaphragm, as several studies showed that an increase in IAP helps to stabilize the spine (Essendrop, Andersen & Schibye, 2002; Cholewicki, Juluru, Radebold, Panjabi & McGill, 1999; Hodges, Eriksson, Shirley & Gandevia, 2005). One explanation for how IAP might cause enhanced stability is by creating an extensor moment that seems to increase spinal stiffness and unload the spine when force is

6

applied (Stokes, Gardner-Morse & Henry, 2010). According to a more famous theory, IAP helps preserving the hoop like geometry around the lumbar spine. Thereby the hip and chest are kept in optimal positions relative to one another, which allows optimal tension and force generation (Kawabata et al., 2014; Hagins, Pietrek, Sheikhzadeh, Nordin & Axen, 2004). However, some studies show that the ability to increase IAP of the diaphragm is position specific (Arjmand & Shirazi-Adl, 2006; Lopes, Nunes, Niza & Dourado, 2016).

Figure 3: IAP producing mechanism of the diaphragm, the pelvic floor and the abdominal muscles (Frank, Kobesova & Kolar, 2013)

Another possible mechanism for the diaphragm to increase spinal stability are its fascial connections. These go to the spinal vertebrae, the thoracolumbar fascia and the pelvic floor muscles (Shirley et al., 2003; Bordoni & Zanier, 2013; Hodges, Sapsford & Pengel, 2007) thereby creating stabilizing forces (Hagins et al., 2004). All of which are influenced by proper diaphragm function.

2.3.2 The effect of 90/90 breathing on stability

The 90/90 bridge with ball and balloon is one example of a breathing technique that actively works on proper diaphragm mechanism and thereby also IAP generation. It will be called 90/90 breathing for convenience reasons throughout the thesis. The 90/90 breathing “was designed to optimize breathing and enhance both posture and stability to improve function and/or decrease pain” (Boyle et al., 2010, p. 1). These effects are supposed to be achieved by improving the zone of apposition (ZOA) of the diaphragm, which is the part of the muscle shaped like a dome (see Figure 4). If the ZOA is decreased the ability of the diaphragm to inhale sufficient air in a correct way is diminished. This affects the diaphragms ability to build up IAP.

7

The transversus abdominis activation also decreases with a smaller ZOA (Boyle et al., 2010), which again affects lumbar stabilization ability (Roussel et al., 2009).

Figure 4: Representation of the zone of apposition (ZOA) in an optimal and sub-optimal shape (Boyle et al., 2010)

However, when the ZOA is optimized the diaphragm can work efficiently. This is achieved by the 90/90 breathing because of the muscles that get activated and the specific body position, which is described in detail in the methods section. Doing the technique, the pelvic floor and the diaphragm get repositioned into a parallel alignment to each other (Boyle et al., 2010), thereby working against the lower and upper crossed syndromes (Boyle et al., 2010; Perri & Halford, 2004). It also works against lumbar extension, rib elevation and external rotation, anterior pelvic tilt and paraspinal activity (Boyle et al., 2010), thereby improving the ZOA. Moreover, the increased activation of the rectus abdominis, transversus abdominis, internal oblique and external oblique muscles during the technique may improve their ability to oppose the diaphragm, increasing the ability to maintain an optimal ZOA (Boyle et al., 2010). The increased activation of the abdominals might also contribute to a more stable spine in general. This is possible since all four are an important part in the concept of spinal stability, or core stability as it has been called over the past two decades.

The optimal ZOA restored with the 90/90 breathing helps the diaphragm fulfill its respiratory and postural dual role. Especially when doing physical activity this optimal functioning of the dual role is important because exercise increases both the demand on the respiratory and stability function of the diaphragm (Boyle et al., 2010; Hodges et al., 1997; Hodges, Heijnen & Gandevia, 2001). If the respiration is dysfunctional the body must give more attention to it,

8

thereby possibly decreasing the stabilizing potential of the diaphragm as well as of other important core muscles, like the transversus abdominis (Hodges & Gandevia, 2000b). Hodges et al. (2001) showed this in their study, where they found that the postural activity of the diaphragm decreases when the respiratory demand increases. Reduced activity in the main breathing muscles might be compensated by other breathing and non-breathing muscles, thereby possibly overloading them (Courtney, 2009).

Even thought, the literature suggests the 90/90 breathing should affect the stability, the author of this thesis found no studies that actually investigated the effect that 90/90 breathing has on stability. This is part of the research gap that will be examined with this thesis.

2.4 Anatomical and functional definition of the core

In the scientific literature the definitions of the core differ in terms of how many and which structures are included. McGill (2010) includes the spine, the abdominal wall muscles, the back-extensor muscles, the quadratus lumborum as well as the latissimus dorsi, the psoas and gluteal muscles. Kibler, Press and Sciascia (2006) include the proximal lower limbs, hips, pelvis, spine and abdominal structures. Behm, Drinkwater, Willardson and Cowley (2010) define it as the axial skeleton including the shoulder and hip girdle and all soft tissues originating from the axial skeleton with proximal attachments. Richardson, Jull, Hodges and Hides (1999) describe it as “a box with the abdominals in the front, paraspinals and gluteals in the back, the diaphragm as the roof, and the pelvic floor and hip girdle musculature as the bottom” (Akuthota & Nadler, 2004, p. 1). 29 pairs of muscles are part of that ‘core box’ (Akuthota, Ferreiro, Moore & Fredericson, 2008). Including the transversus abdominis and the multifidus muscles, that are often stated as specifically important for core stability in the literature (Martuscello et al., 2013; Sharma, 2012; Key, 2013). This definition comes the closest to the concept of the core the 90/90 breathing technique wants to restore. With the diaphragm and pelvic floor as the top and bottom parallel to each other as in a real box and the muscles surrounding the lumbar spine as the wall of the box. All of them get activated with the breathing technique, thereby stabilizing it and creating a rigid cylinder (see figure 5). We will refer to that definition of the core when talking about it in this thesis.

Functionally the central structure of the core is the thoracolumbar fascia, because it is an important connection between several structures within that box (Bordoni & Zanier, 2013). All these connections allow a better formation of the stable hoop around the spine. They also play

9

an important part in the diaphragms role in stability, as mentioned in section 2.3.1. Moreover, by the connections to the latissimus dorsi and gluteus maximus it connects the lower to the upper limbs, thereby making the core the center of the integrated kinetic chain (Kibler et al., 2006; Akuthota & Nadler, 2004). However, the core also fulfills important local functions, namely force generation and even more important stability. This stability is achieved in part by the APAs mentioned in section 2.3.2. Another stability mechanism of the core is when several core muscles become synergists and co-contract to form a stiff muscular corset (McGill, 2010; Borghuis, Hof & Lemmink, 2008; van Dieën Luger & van der Eb, 2012).

Figure 5: Anatomical view of the core-cylinder with the diaphragm as the top, the pelvic floor as the bottom, the paraspinals in the back and the abdominals in the front (Chauhan, n.d.)

2.5 Core stability definition and components

Hodges (2004) defined core stability as “dynamic process of controlling static position in the functional context but allowing the trunk to move with control in other situations” (Waldhelm & Li, 2012, p. 1). Panjabi (1992) defined CS as the active, the passive and the neural and feedback subsystem. Liemohn, Baumgartner and Gagnon (2005) defined CS based on this concept as “the functional integration of the passive spinal column, active spinal muscles, and the neural control unit in a manner that allows the individual to maintain the intervertebral neutral zones within physiologic limits while performing activities of daily living” (Liemohn et al., 2005, p. 1). Grenier and McGill (2007) defined core stability as the spine’s ability to survive perturbations. Kibler et al. (2006) defined it as “the ability to control the position and motion of the trunk over the pelvis to allow optimum production, transfer and control of force and motion to the terminal segment in integrated athletic activities” (Kibler et al., 2006, p. 2). Controlling the trunk over the pelvis keeps the center of mass (COM) over the center of pressure

10

(COP), which is important for balance, especially in single leg activities (Cinar-Medeni, Baltaci, Bayramlar and Yanmis, 2015).

This overall disunity in the literature concerning CS definitions can be explained, besides the wide spread anatomical definitions, by the fact that the exact function of CS and which structures contribute how much is task specific (McGill, Grenier, Kavcic & Cholewicki, 2003).

For this thesis we will refer to CS as a mixture of the definitions from Kibler et al. (2006) and Liemohn et al. (2005). Core stability will be defined as the ability to control the position and motion of the trunk over the pelvis by the functional integration of the passive, active and neural subsystem. For the purpose to maintain the intervertebral neutral zones within physiologic limits during daily and athletic movements as well as the optimum production, transfer and control of force and motion in integrated kinetic chain activities.

This disunity of definitions also explains why there is no consent on what the components of CS are. Some include a combination of neuromuscular control and muscular capacity (Borghuis et al., 2008; Leetun, Ireland, Willson, Ballantyne & Davis, 2004; Butowicz, Ebaugh, Noehren & Silfies, 2016). Muscular capacity consists of muscle strength and endurance, with neuromuscular control being the ability to orchestrate and coordinate both capacities (Borghuis et al., 2008). Of the three components neuromuscular control is of specific importance since it allows the precise coordination of muscle actions at the right time to produce movement and stability (Borghuis et al., 2008). This in the first place makes CS possible and is aided by muscular endurance, strength or both depending on the situational demands on the system. Other researchers include more components as central to the concept of core stability, like flexibility and function (Waldhelm & Li, 2012) or even power and balance (Sharma, 2012). However, for this study we will refer to the components of core stability as the combination of neuromuscular control and muscular capacity.

2.6 Measuring Core stability

With the complexity and disunity of the topic comes a wide variety of different tests that aim on measuring core stability. Depending on the definition the researchers had of CS, they chose tests with different qualities that measure different components of CS. The most widely used test is probably the trunk endurance test battery from McGill, Childs and Liebenson (1999). It consists of 4 different measurements of the anterior, posterior and lateral trunk endurance. However, endurance is only a part of what CS consists of as detailed in the section

11

above. Apart from that a wide variety of tests claim to measure CS ranging from strength tests to neuromuscular control tests and functional tests (Waldhelm & Li, 2012). However, a true gold-standard measurement for CS is missing. This makes the choice for a core stability test difficult. Especially because evidence suggests that important core muscles and stabilizing mechanisms work task specific (McGill et al., 2003; McGill, 2010; Frank et al., 2013). When not researching CS in connection with a certain movement or performance, but the concept of CS at whole, it makes thereby sense to use several tests. A logical approach to choosing the right tests for measuring CS in its complexity is to look at the movement planes the core works in. These are the frontal plane, the sagittal plane and the transversal plane. The core allows and restricts movement in all three planes, making local force production, distal mobility and movement along the kinetic chain possible.

Three tests that fall into these movement planes are the double leg lowering (DLL), the unilateral hip bridge (UHB) and the single leg stand (SLS). In the DLL the legs of a supine lying person move from a vertical to a horizontal position which represents the sagittal plane. The UHB, where a supine lying person with the feet on the floor lifts the hip up and extends one leg once in the neutral position, challenges CS multi-planar (Butowicz et al., 2016). Namely in the sagittal and transversal plane. Lastly, the SLS is a test where the person stands on one leg with the other leg held in the air. Looking at how the feet are lined up in the frontal plane in a normal stand one could argue that a stand on only one leg makes frontal stabilization harder, challenging CS in that plane. Besides their categorization to certain movement planes, all three tests were chosen because they are supposed to measure the neuromuscular control component of CS (Butowicz et al., 2016; Waldhelm & Li, 2012; Sharrock, Cropper, Mostad, Johnson & Malone, 2011). The 90/90 breathing is more likely to affect this component rather than muscular capacity, because it teaches proper stabilization mechanisms and posture.

2.7 Core stability and its influence on injury and performance

2.7.1 Core stability and injury

Looking at the definitions of core stability and the functional connections of the core it makes sense that a badly functioning core could have a wide variety of negative impacts on the body. The spine is specifically prone to these impacts, considering it would buckle under a load as small as 20 Newton (N) or 2 kg without the structures surrounding it creating stability (Panjabi, 1992). One of the most recognized impacts of CS is on lower back pain. Several

12

studies show that LBP is associated with poor control and function of the core muscles and structures (Borghuis et al., 2008; Hubley-Kozey & Vezina, 2002; McGill et al., 2003). And even though it is not clear yet if these factors cause LBP or are caused by LBP, studies show that a functioning and stable core can play an important role in the prevention of LBP (Biering-Sørensen, 1984; McGill, 2015).

Table 1: Overview of the most important literature studying the effect of core stability (CS) or core stability training (CST) on lower back pain(LBP).

Author Title Aim Positive effect

of CS on injury Result Carpes, F. P., Reinehr, F. B., & Mota, C. B. (2008) Effects of a program for trunk strength and stability on pain, low back and pelvis kinematics, and body balance: a pilot study

Research the effect of CST on LBP during gait in women Yes CST decreased LBP and increased postural stability Wang, X. Q., Zheng, J. J., Yu, Z. W., Bi, X., Lou, S. J., Liu, J., ... & Shen, H. M. (2012) A meta-analysis of core stability exercise versus general exercise for chronic low back pain.

To review the effect of CST and general exercise on LBP

Yes CST had better short-term pain relief. No difference after 6/12 months. CST resulted in better functional status Chang, W. D., Lin,

H. Y., & Lai, P. T. (2015)

Core strength training for patients with chronic low back pain

To review the effect of different CST methods on LBP

Yes All CST methods reviewed decreased LBP. More than general resistance training

More evidence has been compiled to show that CS training can be an effective means in the rehabilitation of patients that are already suffering from LBP (Wang et al., 2012; Carpes, Reinehr & Mota, 2008; Shamsi, Rezaei, Zamanlou, Sadeghi & Pourahmadi, 2016). These findings become especially important when looking at the occurrence for LBP with a general prevalence in the USA and Europe ranging from 15-40%, (Carpes et al., 2008; Shamsi et al., 2016). For a summary of the most important literature of CS and LBP see Table 1.

2.7.2 Core stability and performance

When it comes to performance parameters the literature agrees much less on the effect of core stability training or if there even is an effect. A big body of literature has been produced so far on CS training and its effects on several different types of sport and performance

13

variables. Some of the literature shows positive effects of CS training on throwing performance (Sharrock et al., 2011), landing forces in female capoeira athletes (Araujo, Cohen & Hayes, 2015), unipedal postural stability (Lee, You, Kim & Choi, 2015), balance (Sandrey & Mitzel, 2013) and athletic performance (Imai & Kaneoka, 2016; Nesser, Huxel, Tincher & Okada, 2008).

Table 2: Overview over the most important literature studying the effect of core stability (CS) or core stability training (CST) on performance

Author Title Aim Positive effect

of CS on performance Result Sharrock, C., Cropper, J., Mostad, J., Johnson, M., & Malone, T. (2011)

A pilot study of core stability and athletic performance: is there a relationship? Is there a relationship between CS and athletic performance measures? No Only 1 of 4 tests showed a significant but weak correlation with CS performance Schilling, J. F., Murphy, J. C., Bonney, J. R., & Thich, J. L. (2013) Effect of core strength and endurance training on performance in college students: randomized pilot study If CS endurance and strength training have an effect on strength, core endurance and performance tests

Inconsistent CST improved core endurance and partly strength but no performance measurements Araujo, S., Cohen, D., & Hayes, L. (2015)

Six weeks of core stability training improves landing kinetics among female capoeira athletes: a pilot study If CST affects landing kinetics during a drop jump in female capoeira athletes

Yes Peak landing force of landing phase 1 and 2 was reduced after CST. Jump height did not improve Lee, N. G., You, J. S. H., Kim, T. H., & Choi, B. S. (2015) Intensive abdominal drawing-in maneuver after unipedal postural stability in nonathletes with core instability If CST improves unipedal postural stability in non-athletes Yes CST decreased postural sway and eliminated core instability

Imai, A., & Kaneoka, K. (2016)

The relationship between trunk endurance plank tests and athletic performance tests in adolescent soccer players If CS is related to athletic performance tests

Inconsistent Core endurance tests showed positive correlations with the Yo-Yo, Cooper and step 50 agility test but not with any jump or sprint measurements OZMEN, T. (2016) Relationship between core stability, dynamic balance and jumping performance in soccer players If CS is related to dynamic balance and jumping performance in male soccer players No No positive relationship between CS and balance or jump performance was found.

14

However, these findings are not consistent throughout the literature with a different body of studies showing no effect of CS training on performance in balance, soccer, handball or overall athletic performance (McCartney & Forsyth, 2017; Schilling, Murphy, Bonney & Thich, 2013; Nesser & Lee, 2009; Chittibabu, Ramesh Kannan & Jayakumar, 2013).

Several reasons could be the cause of this disunity. First, as mentioned above, core stability seems to be task specific. With the wide variety of different tests claiming to measure CS it might be easy to pick a test that does not capture the components of CS that are mainly active in the studied movement or task. Moreover, it might be easy to test the CS components in a wrong set-up. A second reason might be that a lot of general exercises automatically train the core, especially the ones used in athletic training. Thereby any advantage specific CS training might have over general exercise could be diminished.

As shown, a lot of research exists on the effect of CS training interventions. However, the short-term effects of CS exercises on performance lack research so far. Methods like abdominal bracing (AB) and hollowing have been mainly used as acute CS interventions in the literature and AB has been proven to increase spinal stiffness (Liebenson, Karpowicz, Brown, Howarth & McGill, 2009; Vera-Garcia, Elvira, Brown & McGill, 2007). Studies on the effect of these interventions on performance however are scarce. Apart from these two techniques the author only found one other study on short-term effects of a CS intervention on performance. Kaji, Sasagawa, Kubo and Kanehisa (2010) found positive effects of a CS intervention on a bipedal postural balance task done directly afterwards. For a summary of the most important literature of CS and performance see Table 2.

2.8 Functional Training

The beforehand mentioned research gap about the transient effects of core stability exercises is important when it comes to Functional Training. According to Cosio-Lima, Reynolds, Winder, Paolone and Jones (2003, p. 1) FT is “the ability of the neuromuscular system to perform dynamic concentric, eccentric, and isometric stabilization contractions in response to gravity, ground-reaction forces, and momentum.” This shows that stability is an important part in FT and one method coaches try to achieve this for the dynamic or heavy exercises of the main part in training sessions are core activations. They consist of either general or more training specific exercises that work on the core and its stability. Examples for core activations are plank variations or certain exercises in the quadruped position (Boyle, 2012).

15

Core activations are used as a short-term way to enhance CS, with the intention to make the following exercises more safe and effective because the stability and functionality of the integrated kinetic chain are supposedly improved. However, scientific proof for these claims is almost all but missing. Apart from the study by Kaji et al. (2010) the author of this thesis did not find any other study on the short-term effects of core activations on performance or studies if core activations even improve core stability in the first place.

This research gap leads to the main rationale for conducting this study; to find out if a core activation can really enhance CS. Breathing techniques lend themselves to use as core activations. This is because breathing has a big body of literature showing it could affect CS positively by working on the core muscles, IAP generation as well as causing negative effects for CS and the whole movement function if not working properly. Especially diaphragmatic breathing seems to be important for CS. However, studies concerning short lasting stability enhancements through breathing have not been found. Moreover, in the general public as well as among some practitioners and coaches, breathing is still not recognized for the beneficial effects it can have on performance and disbalances as much as the literature suggests. Therefore, the author chose the diaphragmatic breathing technique 90/90 breathing as a core activation to study whether this strategy of short-term CS enhancement done so frequently in FT actually works, and if it could be done with a breathing technique.

2.9 Aim

The Aim of the study was to test if a short-term activation of the core by doing the 90/90 breathing would impact core stability.

The research question was thereby if it is possible to influence the short-term core stability with an activation of the core muscles by doing the 90/90 breathing.

3. Methods

3.1 Subjects

To test whether 90/90 breathing influences core stability 44 subjects (43% women) were tested. They were all students at a university in the southern part of Sweden and aged between 18 and 35 years. This age span was chosen to test young adults. All subjects were known

16

personally by the author and asked verbally in person or written via the social media platform Facebook. 53 subjects were asked, 9 subjects declined participation out of time reasons or because they fell into one of the exclusion criteria. Exclusion criteria were a BMI over 25, current illness or disease, major injuries like broken bones or torn ligaments within the past 6 months, extended experience in breathing techniques or practice of them on a regular basis like in yoga and doing heavy lifts like squats or deadlifts on a regular basis. The BMI was limited because a study on children showed that those with normal weight had a better core stability than overweight ones (Haggag, 2017). Heavy lifters usually already have developed breathing strategies like the Valsalva maneuver to increase core stability and are thereby less likely to benefit from the short-term effects of 90/90 breathing. People doing yoga or other breathing techniques on a regular basis might already possess a good control over their diaphragm breathing and would be thereby also less likely to benefit from the intervention. “Regular” in this study was defined as performing yoga or heavy lifting at least once a week for longer than a month. Being a heavy lifter meant lifting more than 1.5 times the bodyweight in the squat or deadlift for at least 5 repetitions.

Four subjects were excluded from the data analysis due to fulfilling exclusion criteria which was unknown when inviting them to participate. Therefore, 40 subjects were included in the data analysis.

3.2 Experimental Design

A quantitative study was conducted in the form of an intervention trial to test the research question. The subjects were randomly divided into a control group (CG) and a breathing group (BG) that did the intervention. This was done to be able to tell if possible results of the breathing technique are due to the effect of the intervention or due to repeated measures effects. Moreover, to see if there was a difference between doing the intervention or not. Twenty-one subjects were in the CG and 19 in the BG.

Three tests were chosen to measure core stability, the double leg lowering, the unilateral hip bridge and the single leg stand. One test for each movement plane of the core. They will be described in detail later on. The tests were done in a pre-test post-test manner with the first test round at the beginning of the testing session and the second round at the end. The BG did the breathing intervention in between whereas the CG did nothing but stayed in a similar position to the one the subjects had during the intervention.

17

3.3 Testing procedure

Before the start of the testing each subject was advised not to engage in any sport on the day of the tests and to avoid intense exercise on the day before, especially of the legs and the core. This was done to avoid muscle soreness on the testing day or impaired muscle function during the tests. After arriving the subjects were asked to fill out the informed consent as well as a questionnaire. In the latter they were asked about their weight, height, age, gender as well as about their engagement in sport in hours and times a week and their training and breathing experience.

Next three assessments were performed. The first two were the high-low-test (HLT) and the lateral-rib-expansion-test (LRET) (Nelson, 2012), as can be seen in Figure 6 and Figure 7. Both test the functionality of breathing in different situations. For the HLT the subject stood in front of the examiner and placed one hand on the chest, the other on the lower abdomen. The examiner then observed the movement of the hands caused by the inhalation. As described in the background during quiet breathing the abdomen was supposed to move first as an indicator of functional diaphragmatic breathing. Dysfunctional breathing was indicated by the chest moving first and more dominantly than the abdomen. For the LRET the examiner positioned himself behind the subjects and placed his hands with permission on the side of the lower ribs. The subjects were then instructed to take three deep breaths. If the ribcage expanded lateral and the shoulder area rose only minimal the breathing was functional. With a dominant shoulder movement upwards and or a mainly upwards movement of the lower ribs breathing was classified as dysfunctional. For both tests the subjects were not told beforehand what the examiner was looking for to not influence their natural breathing behavior.

Figure 6: High-low-test starting position. One hand on the chest and one on the abdomen to see which

area dominantly moves during quiet breathing (Nelson, 2012)

Figure 7: LRET-test set-up. The examiner’s hands are on the side of the lower chest to feel the motion of the lower ribs during deep breathing (Nelson, 2012)

18

The third assessment was a kicking test to find out which leg was the more stable dominant leg (DL) one in a single leg stand. For that purpose, the examiner rolled a ball to the subjects and they were instructed to kick it back. Reed, Jennings, Nakamura and Wilson (2015) found that the leg used as standing leg during the ball kicking legs showed longer unipedal standing times compared to the kicking leg and thereby seems to be the more stable leg. After the assessments the subjects randomly chose which group they would belong to by picking one folded paper out of a pile with either control or breathing written on them. The papers were matched for the number of participants so that in the end the same number of subjects would be tested for both groups.

The three tests done in the first testing round (Pre) as well as in the second testing round (Post) were a single leg stand, a unilateral hip bridge and a double leg lowering. These tests were chosen because each test dominantly tested one of the three motion planes the core can move in. In the DLL subjects were lying on a mat on the floor with the examiners hand below the L4-5 area of the lumbar spine and the legs as straight as possible vertically in the air. The subjects then lowered them in a steady speed taking 10 seconds (s) from the top position all the way down (see Figure 10). This was counted down out loud by the examiner. The procedure was shown to the subjects beforehand and the examiner made sure they understood how to do it. Before the DLL started the subjects were shown how to do a posterior pelvic tilt to preserve the natural form of the spine during the exercise. They were told to hold that posterior pelvic tilt by pressing the lumbar spine into the ground throughout the DLL. Once in the start position a wooden stick with a digital inclinometer (Model MDP01, Shenzen Temie Technology Co. Ltd., Shenzhen, China) attached to it was placed alongside the femur to measure the angle in degrees the participants were able to achieve. The test was stopped when the subjects could not hold the posterior pelvic tilt anymore and the spine started lifting off the ground (Krause, Youdas, Hollman & Smith, 2005). Two repetitions of the DLL were performed, and the best try was taken for the analysis. The DLL represented the test for the sagittal movement plane of the core. For the UHB the subjects were instructed to lie in a supine position on a mat with their feet on the floor in a position where their fingers could still touch their heels. A belt (Hip Belt, Exxentric AB, Bromma, Sweden) was placed around their hip with a wooden stick plus the digital inclinometer attached to it. The subjects were instructed to cross the arms on the chest, lift up their hip into a normal hip bridge and once in the neutral position extend one leg while trying

19

to keep the hip as stable in the neutral position as possible (see Figure 9). Time was taken for how long the participants could hold the position. The examiner placed a goniometer (12 Inch plastic, 66fit, West Pinchbeck, United Kingdom) fixed to a 10° angle on the joint axis of the knee to track the sagittal movement of the hip while the inclinometer measured the transversal movement of the hip. Once the movement of the hip broke 10° in one of both planes the time was stopped (Butowicz et al., 2016). The test was first done with the dominant leg and repeated with the non-dominant leg (NDL). The UHB represented the test for the transversal movement plane of the core. For the SLS the subjects were instructed to step on a force plate (Model PJB-101, Advanced Mechanical Technology, Inc., Watertown, USA) with their stable foot, lifting up the other leg with the knee and the heel in front of the standing leg and the arms across the chest (see Figure 8). Once in position subjects were supposed to find a stable stand and close the eyes on the command of the examiner while trying to stand as stable as possible for 10 seconds (Lee et al., 2015). They were given one trial round and were tested three times after with 30 seconds pause in between each repetition. When the subjects fell over or started twisting their foot away from the starting point the try was repeated. The center of pressure in its maximum range was measured in X- and Y-direction on the force plate and used as test variable. The test was done barefoot.

Figure 8: Set-up of the unilateral hip bridge(UHB) with the digital Inclinometer attached to a hip belt via a wooden stick.

Figure 10: Set-up of the double leg lowering (DLL) with the examiner placing the digital inclinometer attached to a wooden stick along the femur

Figure 9: Set-up of the single leg stand (SLS) on a force plate.

20

The breathing group then performed the 90/90 breathing detailed below after a pause of 5 minutes. These 5 minutes were used to explain the technique to the subjects and let them try it under the supervision of the examiner, so he could correct them and make sure they did it right for the actual exercise. All subjects of the BG were able to perform the breathing after the explanation. The intervention itself took 3 minutes. The control group did nothing during these 8 minutes, but subjects were instructed to go into similar positions as the BG to exclude the possibility that any other movement could have had an effect on their performance. Therefore, they sat for the first 2.5 minutes before lying in a supine position on the mat with the feet on the floor for the next 5.5 minutes.

In the Post-test the three tests were performed in a random order. This was done so that no specific order of the tests could have an impact on the effect of the intervention. For this purpose, the name of each test was written on a paper that was folded with the subject picking a random one before each test.

3.4 Breathing intervention

The 90/90 breathing was done as described in detail in the paper of Boyle et al. (2010). Subjects lay in a supine position on a mat, the feet against a wall with 90° in the knees and hip, a soft ball between the thighs. One hand was holding a balloon on the lips, the other hand was held overhead (see Figure 12). Subjects were instructed to inhale through the nose via diaphragmatic breathing into the abdomen and lower ribs, then exhale into the balloon. During the exhale the feet were pushed into the wall and pulled down isometrically with moderate force. Moreover, the legs pushed the ball together during the exhale.

21

This tension was constantly held for one set. After the exhale a one second pause followed and then the inhale during which the air was supposed to stay in the balloon without pinching it. The posture of the core created with the exhalation was also supposed to be kept. After four breaths into the balloon subjects released the air of the balloon and relaxed. This was one set and four of them were performed with short pauses in between, only as long until the subjects felt ready to go for the next set. After two sets the setup of the arms was switched.

3.5 Ethical and Social Considerations

The study will follow the guidelines of the Declaration of Helsinki. None of the tests done in this study involved invasive techniques. The core stability tests wielded, as every physical intervention an existing, albeit very small injury risk. The breathing itself could induce dizziness because of the high volume of inhaled air and thereby a high amount of oxygen in a short time, to which subjects might have not been used to. This was clearly communicated to the subjects as well as their freedom to end the participation whenever they wanted without having to justify themselves.

From a social perspective the study did wield some advantages for the general population. The breathing technique could help people with lower back pain or neck pain as described in the background. This benefit would be even more prominent in certain situations when the 90/90 breathing would show to increase CS, which works against LBP as shown in the background. For recreational active people, but also sedentary people, the knowledge gained could help reduce LBP also by increasing the awareness for the 90/90 breathing.

Another benefit would be that the knowledge gained could help make training and preparation more effective, by improving the warm-up and thereby possibly enhancing performance or making training and activity safer. It would be especially interesting for the Functional Training community, since the exercises and methods like the ones used in this study are common practice there.

3.6 Statistical Analysis

All data analyses were conducted using Statistical Package for the Social Sciences (SPSS v24, Chicago, United States of America). The data was first tested for normal distribution with the Shapiro-Wilk test. It showed normal distribution for the double leg lowering data and not

22

normal distribution for the unilateral hip bridge and center of pressure data. Therefore, all calculations with DLL variables were done with parametric tests whereas all calculations with UHB and COP variables were done with none-parametric tests.

An Independent t-test and Mann-Whitney U tests were done to test for differences in the test performance between CG and BG in the Pre- and Post-test. This was the first test series. The same tests were used for the calculation for differences in the performance change from Pre to Post between groups. This was the third test series. A Paired Samples t-test and Wilcoxon tests were done to test for performance improvements from Pre to Post within the groups. This was the second test series. To check for any group differences between the dominant leg and the non-dominant leg in the UHB at Pre or Post a Wilcoxon test was calculated in addition to the second test series. The significance level was set at p≤ 0.05 for all tests.

A secondary analysis to check for group differences in all CS variables in several sub-categories of the population was also done. This had the aim to find out if any particular group of the subjects had specific influence on the results. Sub-categories were gender, breathing pattern and time spent doing sports per week in hours (TSDS). For gender the subjects were divided into men and women. The statistical calculations were done with an Independent t-test for the DLL data and Mann-Whitney U tests for the UHB and COP data. For breathing pattern, the subjects were divided into three categories, functional breathing, partly dysfunctional breathing and dysfunctional breathing. The calculation was done with an ANOVA. Moreover, a Spearman correlation between all CS variables and the TSDS was calculated. The strength of the correlation was evaluated after the study of Hinkle, Wiersma and Jurs (2003), with an r=0.0 to 0.2 being negligible, r=0.2 to 0.4 being low, r=0.4 to 0.6 being moderate, r=0.6 to 0.8 being high and r>0.8 being excellent.

4. Results

4.1 Descriptive statistics

The descriptive statistics of the most important subject variables can be seen in Table 3. According to the two tests for breathing functionality, the high-low-test and the lateral-rib-expansion-test, 10 people had completely functional breathing, 15 had partly dysfunctional and 15 had completely dysfunctional breathing. 34 subjects had gym experience to some degree, 20 had experience in team sports, namely soccer, volleyball, basketball, American football, rugby

23

and hurling and 9 had experience in individual sports, namely running, high jumping, thai boxing, swimming, cycling, tennis, gymnastics and water skiing.

Table 3: Descriptive Subject Data

Number Age (y) Weight

(kg) Height (cm) TSDS (h) SSPW YTE Full population 40 23.3 70.5 175.9 5.4 3.5 6.5 Women 17 21.87 62.73 168.13 4.50 3.57 5.03 Men 23 24.08 75.16 180.56 5.94 3.54 7.32 CG 21 23.2 71.9 176.3 6.05 3.95 7.14 BG 19 23.3 69.0 175.4 4.68 3.11 5.76

y = years, kg = kilogram, cm = centimeters, h = hours, TSDS = Time spent doing sports per week, SSPW = Sport sessions per week, YTE = Years of training experience

4.2 Main effect calculations

4.2.1 Group difference at the first and second test round

The first series of tests done was to see if there were differences between the control group and the breathing group for Pre or Post. No significant differences were found between the groups at baseline (see Table 4). However, the UHB data at Post showed a significant difference for the dominant leg (p=0.029) with a mean holding time of 42.11s for the BG and 27.24s for the CG as well as for the non-dominant leg (p=0.031) with a mean holding time of 51.51s for the BG and 30.35s for the CG (see Table 5). No significant difference between groups was found in the Retest for the DLL data as well as for COP data in X- and Y-direction (see Table 5).

Table 4: Group differences between breathing group and control group at the first test round.

PRE Mean Value BG Mean Value CG p-values

UHB-DL- Pre (s) 36.28 ± 14.73 a _{29.61 ± 18.93 0.233}

UHB-NDL- Pre (s) 44.87 ± 25.52 31.80 ± 25.96 0.091

DLL- Pre (°) 52.82 ± 17.83 59.31 ± 17.00 0.246

COP-X- Pre (mm) 66.82 ± 22.34 62.98 ± 19.94 0.336

COP-Y- Pre (mm) 42.86 ± 7.86 45.96 ± 11.98 0.924

BG = breathing group, CG = control group, UHB = unilateral hip bridge, DLL = double leg lowering, COP = center of pressure, DL = dominant leg, NDL = non-dominant leg, Pre = first test round, a _{= standard deviation}

24

Table 5: Group differences between breathing group and control group at the second test round.

POST Mean Value BG Mean Value CG p-values

UHB-DL-Post (s) 42.11 ± 21.71 a _{27.42 ± 17.47 0.029*}

UHB-NDL-Post (s) 51.52 ± 31.09 30.35 ± 25.49 0.031*

DLL-Post (°) 61.39 ± 19.75 57.30 ±15.70 0.470

COP-X-Post (mm) 61.83 ± 34.02 52.27 ± 11.49 0.473

COP-Y-Post (mm) 42.79 ± 10.14 41.68 ± 6.06 0.860

BG = breathing group, CG = control group, UHB = unilateral hip bridge, DLL = double leg lowering, COP = center of pressure, DL = dominant leg, NDL = non-dominant leg, Post = second test round, * = significant at p < 0.05, a _{= standard deviation}

4.2.2 Performance change within groups from the first to second test round

The second series of tests was conducted to see if there was a change in performance within the groups from Pre to Post (see Table 6). The data for the UHB showed no difference in the CG in both legs and in the BG for the dominant leg but a significant difference in the BG for the non-dominant leg (p=0.02) with a mean holding time of 38.01s for Pre (see Table 6) and 40.40s for Post (see Table 6). The data for the DLL showed no difference from Pre to Post for the CG but a significant difference for the BG (p=0.001) with a mean leg lowering distance of 52.82° for Test (see Table 6) and 61.39° for Retest (see Table 6). The COP data showed no significant difference for the BG in X- or Y-direction and no significant difference for the CG in Y-direction. But the data showed a significant difference from Pre to Post for the CG in X-direction (p=0.011) with a mean range of 62.98mm for Pre (see Table 6) and 52.27mm for Post (see Table 6).

The additional test for group differences between the dominant and non-dominant leg in the UHB showed no significant difference between both legs at Pre (p=.136) and Post (p=.243) (see Table 7).

25

Table 6: Performance change within groups from Pre to Post

Mean Value Pre BG Mean Value Post BG p-values Pre-Post Performance Change BG Mean Value Pre CG Mean Value Post CG p-values Pre-Post Performance Change CG UHB-DL 36.28 ± 14.73 a 42.11 ± 21.71 0.227 29.61 ± 18.93 27.42 ± 17.47 0.067 UHB-NDL 44.87 ± 25.52 51.52 ± 31.09 0.020* 31.80 ± 25.96 30.35 ± 25.49 0.689 DLL 52.82 ± 17.83 61.39 ± 19.75 0.001** 59.31 ± 17.00 57.30 ± 15.70 0.336 COP-X 66.82 ± 22.34 61.83 ± 34.02 0.198 62.98 ± 19.94 52.27 ± 11.49 0.011* COP-Y 42.86 ± 7.86 42.79 ± 10.14 0.601 45.96 ± 11.98 41.68 ± 6.06 0.170

BG = breathing group, CG = control group, UHB = unilateral hip bridge, DLL = double leg lowering, COP = center of pressure, DL = dominant leg, NDL = non-dominant leg, Pre = first test round, Post = second test round, * = significant at p < 0.05, ** = significant at p < 0.01, a _{= standard deviation}

Table 7: Mean values and p-values between the legs of the unilateral hip bridge at the first and second test round.

Mean Value DL Mean Value NDL p-values

Difference at Pre (s) 36.28 44.87 0.136

Difference at Post (s)

42.11 51.52 0.243

Pre = first test round, Post = second test round, DL = dominant leg, NDL = non-dominant leg

4.2.3 Difference in the performance change between groups

The third series of tests was calculated to see if there was a difference in the performance change from Pre to Post between groups (see Table 8). The data for the UHB showed no group difference in the performance change for the dominant leg or the non-dominant leg. The DLL data showed a significant difference between groups in the performance change (p=0.001) with a mean difference of -2.01° from Test to Retest in the CG and 8.58° in the BG (see Table 8). The COP data showed no difference in the performance change between groups in X- or Y-direction (see Table 8).

26

Table 8: Difference in the performance change between groups

Mean Value Performance Change Difference CG Mean Value Performance Change Difference BG p-values Performance Change Difference CG-BG UHB-DL -2.37 ± 10.96 a _{5.83 ± 19.84 0.330} UHB-NDL -1.45 ± 16.30 6.64 ± 16.88 0.088 DLL -2.01 ± 9.35 8.58 ± 9.90 0.001** COP-X 1.66 ± 10.65 -0.94 ± 25.14 0.456 COP-Y 0.72 ± 5.73 0.88 ± 8.67 0.440

BG = breathing group, CG = control group, UHB = unilateral hip bridge, DLL = double leg lowering, COP = center of pressure, DL = dominant leg, NDL = non-dominant leg, Pre = first test round, Post = second test round, ** = significant at p < 0.01, a ₌

standard deviation

4.3 Secondary calculations

After these calculations for the main effects secondary tests were done to see if any of the gathered questionnaire variables had an influence on the results. First all the variables were tested for gender differences. Therefore, it was calculated if the data of any test showed group differences between men and women for Pre or Post (see Table 9). No difference between men and women was found for the UHB and DLL in Pre or Post. For the COP data there was no difference found for Pre in X- or Y-direction and for Post in X-direction but a significant difference between genders in Y-direction was found (p=0.002) with men having a mean COP range of 44.61mm and women having a mean range of 38.20mm.

27

Table 9: Group differences between men and women for all test variables at the first and second test round.

Mean Value Men Mean Value

Women p-values UHB-DL-Pre (s) 34.07 ± 18.48 a _{30.63 ± 15.15 0.468} UHB-NDL- Pre (s) 39.82 ± 28.68 35.00 ± 22.28 0.706 UHB-DL-Post (s) 34.62 ± 20.53 33.77 ± 21.84 0.867 UHB-NDL-Post (s) 43.25 ± 33.83 35.67 ± 22.12 0.665 DLL- Pre (°) 57.78 ± 14.65 53.31 ± 21.66 0.421 DLL-Post (°) 60.44 ± 13.59 57.25 ± 23.29 0.634 COP-X- Pre (mm) 66.34 ± 24.07 62.23 ± 14.70 0.922 COP-Y- Pre (mm) 46.43 ± 12.16 41.24 ± 4.47 0.067 COP-X-Post (mm) 59.52 ± 28.63 52.25 ± 17.37 0.645 COP-Y-Post (mm) 44.61 ± 7.97 38.24 ± 7.02 0.002**

UHB = unilateral hip bridge, DLL = double leg lowering, COP = center of pressure, DL = dominant leg, NDL = non-dominant leg, Pre = first test round, Post = second test round, ** = significant at p < 0.01, a _{= standard deviation}

Next the results were checked for the influence of the breathing pattern. Therefore, subjects were divided into three groups. Those with functional breathing (FB) that had no dysfunction in either of both breathing tests. Those with partly dysfunctional breathing (PDB) that had a dysfunction in one of the two breathing tests. And those with dysfunctional breathing (DB) that had a dysfunction in both breathing tests. No group differences between breathing patterns have been found for any of the CS test data (see Table 10).

28

Table 10 Group differences between functional breathing, partly dysfunctional breathing and dysfunctional breathing.

Mean Value FB Mean Value PDB Mean Value DB p-values UHB-DL- Pre (s) 27.14 ± 18.32 a _{39.43 ± 18.06 29.07 ± 13.99 0.175} UHB-NDL- Pre (s) 51.89 ± 33.19 38.35 ± 25.65 29.33 ± 19.56 0.254 UHB-DL-Post (s) 41.57 ± 26.40 36.94 ± 22.28 27.14 ± 13.14 0.468 UHB-NDL-Post (s) 58.99 ± 39.31 36.76 ± 27.64 33.14 ± 22.24 0.240 DLL- Pre (°) 52.16 ± 8.12 62.78 ± 20.74 51.67 ± 16.47 0.154 DLL-Post (°) 55.87 ± 11.22 63.98 ± 19.23 56.22 ± 18.84 0.392 COP-X- Pre (mm) 59.56 ± 26.70 65.86 ± 16.39 66.82 ± 22.40 0.496 COP-Y- Pre (mm) 41.58 ± 9.19 47.98 ± 13.12 42.70 ± 6.10 0.619 COP-X-Post (mm) 58.56 ± 16.69 51.27 ± 12.62 61.63 ± 36.75 0.736 COP-Y-Post (mm) 45.17 ± 7.74 40.99 ± 5.61 41.73 ± 10.49 0.343

UHB = unilateral hip bridge, DLL = double leg lowering, COP = center of pressure, DL = dominant leg, NDL = non-dominant leg, Pre = first test round, Post = second test round, FB = functional breathing, PDB = partly dysfunctional breathing, DB = dysfunctional breathing, a _{= standard deviation}

Last it was calculated whether the time spent per week doing sports (TSDS) was correlated with the CS-tests. The TSDS showed a significant correlation with low strength for the UHB in the dominant leg at Pre (p=.021, r=.364) as can be seen in Figure 12.